ConceptAI & ML
Autoregressive models
A model that generates a sequence by predicting the next element based on the previous elements in the sequence.
Know first
Named as a prerequisite by the lectures that define this. The number is how many say so. A grey one has no page yet.
Opens up
Concepts that name this one as a prerequisite.
Lecturers
2Each in their own terms, from the easiest level.
Phillip Isola
Conditional models by default that predict the next token given previous tokens in a sequence.
Lectures that assume it
10- Let's Build GPT: From Scratch, in Code, Spelled Out
- Building makemore Part 5: Hierarchical Dilated Causal Convolutions (WaveNet)
- Time Series Analysis III
- Intuitions on Language Models and the History of Transformer Architectures
- Reinforcement Learning from Human Feedback: Algorithms & Applications
- Inference Methods for Deep Learning
- An Observation on Generalization and Unsupervised Compression
- Time Series Analysis II
- Volatility Modeling
- Transfer Learning: Data
Connections
24The lectures that define it or assume it, and the concepts next to it.
Concepts10
- Probability
- Sequence modeling
- Conditional probability
- Generative model
- Generative modeling
- Autoregressive models as conditional models
Show 4 moreShow fewer
Lectures14
- Architectures: Memory
- Generative Models: Conditional Models
- Architectures: Transformers
- Generative Models: Basics
- Let's Build GPT: From Scratch, in Code, Spelled Out
- Building makemore Part 5: Hierarchical Dilated Causal Convolutions (WaveNet)
Show 8 moreShow fewer
- Time Series Analysis III
- Intuitions on Language Models and the History of Transformer Architectures
- Reinforcement Learning from Human Feedback: Algorithms & Applications
- Inference Methods for Deep Learning
- An Observation on Generalization and Unsupervised Compression
- Time Series Analysis II
- Volatility Modeling
- Transfer Learning: Data